MCP-1 controls IL-17-promoted monocyte migration and M1 polarization in osteoarthritis.

Osteoarthritis (OA) is a low-grade inflammatory joint illness in which monocytes migrate and infiltrate synovial tissue, differentiating into the pro-inflammatory M1 macrophage phenotype. IL-17 is a proinflammatory mediator principally generated by Th17 cells, which is elevated in OA patients; nevertheless, investigators have yet to elucidate the function of IL-17 in M1 polarization during OA development. Our analysis of clinical tissues and results from the open online dataset discovered that the level of M1 macrophage markers is elevated in human OA tissue samples than in normal tissue. High-throughput screening demonstrated that MCP-1 is a potential candidate factor after IL-17 treatment in OA synovial fibroblasts (OASFs). Immunohistochemistry data revealed that the level of MCP-1 is higher in humans and mice with OA than in normal tissues. IL-17 stimulation facilitates MCP-1-dependent macrophage polarization to the M1 phenotype. It also appears that IL-17 enhances MCP-1 synthesis in human OASFs, enhancing monocyte migration via the JAK and STAT3 signaling cascades. Our findings indicate the IL-17/MCP-1 axis as a novel strategy for the remedy of OA.

The stimulation of exosome generation by visfatin polarizes M2 macrophages and enhances the motility of chondrosarcoma.

Chondrosarcoma is a malignant bone tumor that arises from abnormalities in cartilaginous tissue and is associated with lung metastases. Extracellular vesicles called exosomes are primarily used as mediators of intercellular signal transmission to control tumor metastasis. Visfatin is an adipokine reported to enhance tumor metastasis, but its relationship with exosome generation in chondrosarcoma motility remains undetermined. Our results found that overexpressing visfatin augments the production of exosomes from chondrosarcoma cells. Visfatin-treated chondrosarcoma exosomes educate macrophage polarization towards the M2 but not M1 phenotype. Interestingly, M2 macrophages polarized by exosomes return to chondrosarcoma cells to facilitate cell motility. According to these findings, chondrosarcoma cells emit more exosomes when treated with visfatin. The stimulation of exosome generation by visfatin polarizes M2 macrophages and enhances the motility of chondrosarcoma.

NGF facilitates ICAM-1-dependent monocyte adhesion and M1 macrophage polarization in rheumatoid arthritis.

Rheumatoid arthritis (RA) is a chronic autoimmune inflammatory disorder in which monocytes adhering to synovial tissue differentiate into the pro-inflammatory M1 macrophage phenotype. Nerve growth factors (NGF) referred to as neurotrophins have been associated with inflammatory events; however, researchers have yet to elucidate the role of NGF in RA. Our examination of clinical tissue samples and analysis of data sourced from the Gene Expression Omnibus dataset unveiled elevated expression levels of M1 macrophage markers in human RA synovial tissue samples compared to normal tissue, with no such distinction observed for M2 markers. Furthermore, immunofluorescence data depicted increased expression levels of NGF and M1 macrophages in RA mice in contrast to normal mice. It appears that NGF stimulation facilitates macrophage polarization from the M0 to the M1 phenotype. It also appears that NGF promotes ICAM-1 production in human RA synovial fibroblasts, which enhances monocyte adhesion through the TrkA, MEK/ERK, and AP-1 signaling cascades. Our findings indicate NGF/TrkA axis as a novel target for the treatment of RA.

IL-17 promotes IL-18 production via the MEK/ERK/miR-4492 axis in osteoarthritis synovial fibroblasts.

The concept of osteoarthritis (OA) as a low-grade inflammatory joint disorder has been widely accepted. Many inflammatory mediators are implicated in the pathogenesis of OA. Interleukin (IL)-18 is a pleiotropic cytokine with versatile cellular functions that are pathogenetically important in immune responses, as well as autoimmune, inflammatory, and infectious diseases. IL-17, a proinflammatory cytokine mainly secreted by Th17 cells, is upregulated in OA patients. However, the role of IL-17 in OA progression is unclear. The synovial tissues collected from healthy donors and OA patients were used to detect the expression level of IL-18 by IHC stain. The OA synovial fibroblasts (OASFs) were incubated with recombinant IL-17 and subjected to Western blot, qPCR, and ELISA to examine IL-18 expression level. The chemical inhibitors and siRNAs which targeted signal pathways were used to investigate signal pathways involved in IL-17-induced IL-18 expression. The microRNAs which participated IL-18 expression were surveyed with online databases miRWalk and miRDB, followed by validation with qPCR. This study revealed significantly higher levels of IL-18 expression in synovial tissue from OA patients compared with healthy controls, as well as increased IL-18 expression in OASFs from rats with severe OA. In vitro findings indicated that IL-17 dose-dependently promoted IL-18 production in OASFs. Molecular investigations revealed that the MEK/ERK/miR-4492 axis stimulated IL-18 production when OASFs were treated with IL-17. This study provides novel insights into the role of IL-17 in the pathogenesis of OA, which may help to inform OA treatment in the future.

Glutamine metabolism controls amphiregulin-facilitated chemoresistance to cisplatin in human chondrosarcoma.

Chondrosarcoma is the second most common type of bone cancer. At present, the most effective clinical course of action is surgical resection. Cisplatin is the chemotherapeutic medication most widely used for the treatment of chondrosarcoma; however, its effectiveness is severely hampered by drug resistance. In the current study, we compared cisplatin-resistant chondrosarcoma SW1353 cells with their parental cells via RNA sequencing. Our analysis revealed that glutamine metabolism is highly activated in resistant cells but glucose metabolism is not. Amphiregulin (AR), a ligand of the epidermal growth factor receptor, enhances glutamine metabolism and supports cisplatin resistance in human chondrosarcoma by promoting NADPH production and inhibiting reactive oxygen species (ROS) accumulation. The MEK, ERK, and NrF2 signaling pathways were shown to regulate AR-mediated alanine-serine-cysteine transporter 2 (ASCT2; also called SLC1A5) and glutaminase (GLS) expression as well as glutamine metabolism in cisplatin-resistant chondrosarcoma. The knockdown of AR expression in cisplatin-resistant chondrosarcoma cells was shown to reduce the expression of SLC1A5 and GLS in vivo. These results indicate that AR and glutamine metabolism are worth pursuing as therapeutic targets in dealing with cisplatin-resistant human chondrosarcoma.

Hypoxia-regulated exosomes mediate M2 macrophage polarization and promote metastasis in chondrosarcoma.

Chondrosarcoma is a primary malignant bone tumor. Traditional therapy is not very effective, and it is prone to metastasis in the late stage. The tumor microenvironment (TME) plays a key role in the progression and metastasis of chondrosarcoma, and hypoxia is one of the key factors in the formation of TME. However, the detailed mechanism of how hypoxia affects tumor progression and metastasis in chondrosarcoma is still not fully understood. In this study, we focused on the mechanism of interaction between hypoxic chondrosarcoma cells (SW1353) and macrophages. Our results suggest that hypoxia enhances the release of exosomes from chondrosarcoma cells. These hypoxia-induced exosomes promoted macrophage polarization towards the M2 phenotype, characterized by the expression of CD163 and CD206, but not the M1 phenotype, characterized by CD86 expression. Further analysis revealed that M2 macrophages polarized by exosomes expressed arginase-1 and feedback to chondrosarcoma cells to promote migration. These results suggest that chondrosarcoma cells secrete more exosomes in a hypoxic microenvironment, and these hypoxia-derived exosomes induce the polarization of macrophages into an M2 phenotype, ultimately promoting the metastatic behavior of chondrosarcoma cells.

An approach to the diagnosis of lumbar disc herniation using deep learning models.

Background: In magnetic resonance imaging (MRI), lumbar disc herniation (LDH) detection is challenging due to the various shapes, sizes, angles, and regions associated with bulges, protrusions, extrusions, and sequestrations. Lumbar abnormalities in MRI can be detected automatically by using deep learning methods. As deep learning models gain recognition, they may assist in diagnosing LDH with MRI images and provide initial interpretation in clinical settings. YOU ONLY LOOK ONCE (YOLO) model series are often used to train deep learning algorithms for real-time biomedical image detection and prediction. This study aims to confirm which YOLO models (YOLOv5, YOLOv6, and YOLOv7) perform well in detecting LDH in different regions of the lumbar intervertebral disc. Materials and methods: The methodology involves several steps, including converting DICOM images to JPEG, reviewing and selecting MRI slices for labeling and augmentation using ROBOFLOW, and constructing YOLOv5x, YOLOv6, and YOLOv7 models based on the dataset. The training dataset was combined with the radiologist's labeling and annotation, and then the deep learning models were trained using the training/validation dataset. Results: Our result showed that the 550-dataset with augmentation (AUG) or without augmentation (non-AUG) in YOLOv5x generates satisfactory training performance in LDH detection. The AUG dataset overall performance provides slightly higher accuracy than the non-AUG. YOLOv5x showed the highest performance with 89.30% mAP compared to YOLOv6, and YOLOv7. Also, YOLOv5x in non-AUG dataset showed the balance LDH region detections in L2-L3, L3-L4, L4-L5, and L5-S1 with above 90%. And this illustrates the competitiveness of using non-AUG dataset to detect LDH. Conclusion: Using YOLOv5x and the 550 augmented dataset, LDH can be detected with promising both in non-AUG and AUG dataset. By utilizing the most appropriate YOLO model, clinicians have a greater chance of diagnosing LDH early and preventing adverse effects for their patients.

Melatonin inhibits chondrosarcoma cell proliferation and metastasis by enhancing miR-520f-3p production and suppressing MMP7 expression.

Chondrosarcoma has a high propensity to metastasize and responds poorly to chemotherapy and radiation treatment. The enzymatic activity of matrix metalloproteinases (MMPs) is very important in chondrosarcoma metastasis. Melatonin exhibits anticarcinogenic activity in many types of cancers by suppressing the expression of certain MMP family members, but this has not yet been clearly determined in chondrosarcoma. Our study demonstrates that MMP7 plays an essential role in chondrosarcoma cell proliferation, migration, and anoikis resistance. We also found that MMP7 is highly expressed in chondrosarcomas. Our in vitro and in vivo investigations show that melatonin strongly inhibits chondrosarcoma cell proliferation, migration, and anoikis resistance by directly suppressing MMP7 expression. Melatonin reduced MMP7 synthesis by promoting levels of miR-520f-3p expression, which were downregulated in human chondrosarcoma tissue samples. Pharmacological inhibition of miR-520f-3p markedly reversed the effects of melatonin upon chondrosarcoma proliferation and metastasis. Thus, our study suggests that melatonin has therapeutic potential for reducing the tumorigenesis and metastatic potential of chondrosarcoma via the miR-520f-3p/MMP7 axis.

FDD: a deep learning-based steel defect detectors.

Surface defects are a common issue that affects product quality in the industrial manufacturing process. Many companies put a lot of effort into developing automated inspection systems to handle this issue. In this work, we propose a novel deep learning-based surface defect inspection system called the forceful steel defect detector (FDD), especially for steel surface defect detection. Our model adopts the state-of-the-art cascade R-CNN as the baseline architecture and improves it with the deformable convolution and the deformable RoI pooling to adapt to the geometric shape of defects. Besides, our model adopts the guided anchoring region proposal to generate bounding boxes with higher accuracies. Moreover, to enrich the point of view of input images, we propose the random scaling and the ultimate scaling techniques in the training and inference process, respectively. The experimental studies on the Severstal steel dataset, NEU steel dataset, and DAGM dataset demonstrate that our proposed model effectively improved the detection accuracy in terms of the average recall (AR) and the mean average precision (mAP) compared to state-of-the-art defect detection methods. We expect our innovation to accelerate the automation of industrial manufacturing process by increasing the productivity and by sustaining high product qualities.

"Fragment Width Ratio" as a Predictor of Nonunion for Femoral Shaft Fracture With Third Fragments.

Previous studies have reported that large fracture fragment with displacement might cause nonunion of femoral shaft fractures. We therefore intended to delineate significant risk factors for developing a nonunion predisposed by a major fracture fragment. We analyzed 61 patients who were operated on using interlocking nails for femoral shaft fractures from 2009 to 2018. We classified patients with modified Radiographic Union Scale for Tibia fractures scores of less than 11 or needing reoperations by 1 year postoperatively as nonunion. We thereafter measured parameters of the displaced fracture fragment and fracture site to identify the significant difference between the union and non-union groups. We also applied the receiver operating characteristic curve to demonstrate a threshold value for the fragment width (FW) ratio. Among 61 patients with complete follow-up, no significant difference was found regarding length, displacement, and angulation of fragments between patients with and without union. Except for higher mean FW (P=.03) and the FW ratio (P=.01) in patients with nonunion, the logistic regression analysis demonstrated that FW ratio significantly affected union (P=.018; odds ratio, 0.21; 95% CI, 0.001-0.522). Although a fracture fragment greater than 4 cm with displacement greater than 2 cm was reported to significantly cause nonunions, our study showed that an FW ratio greater than 0.55 instead of fragment size or displacement was predictive for the occurrence of nonunion adjoining to the fracture site. Fixation of the third fracture fragment should not be ignored for preventing a nonunion. More attention should be paid to achieve a better fixation for a major fracture fragment with an FW ratio greater than 0.55 to avoid the development of non-union following the use of interlocking nail for femoral shaft fracture. [Orthopedics. 2023;46(3):169-174.].

MicroRNA-631 Resensitizes Doxorubicin-Resistant Chondrosarcoma Cells by Targeting Apelin.

Chondrosarcoma is the second most common type of bone cancer. Surgical resection is the best choice for clinical treatment. High-grade chondrosarcoma is destructive and is more possible to metastasis, which is difficult to remove using surgery. Doxorubicin (Dox) is the most commonly used chemotherapy drug in the clinical setting; however, drug resistance is a major obstacle to effective treatment. In the present study, we compared Dox-resistant SW1353 cells to their parental cells using RNA sequencing (RNA-Seq). We found that the apelin (APLN) pathway was highly activated in resistant cells. In addition, tissue array analysis also showed that APLN was higher in high-grade tissues compared to low-grade tissues. APLN is a member of the adipokine family, which is a novel secreted peptide with multifunctional and biological activities. Previously, studies have shown that inhibition of the APLN axis may have a therapeutic benefit in cancers. However, the role of APLN in chondrosarcoma is completely unclear, and no related studies have been reported. During in vitro experiments, APLN was also observed to be highly expressed and secreted in Dox-resistant cells. Once APLN was knocked down, it could effectively improve its sensitivity to Dox. We also explored possible upstream regulatory microRNAs (miRNAs) of APLN through bioinformatics tools and the results disclosed that miR-631 was the most likely regulator of APLN. Furthermore, the expression of miR-631 was lower in the resistant cells, but overexpression of miR-631 in the Dox-resistant cell lines significantly increased the Dox sensitivity. These results were also observed in another chondrosarcoma cell line, JJ012 cells. Taken together, these findings will provide rationale for the development of drug resistance biomarkers and therapeutic strategies for APLN pathway inhibitors to improve the survival of patients with chondrosarcoma.

Hierarchical Image Transformation and Multi-Level Features for Anomaly Defect Detection.

Anomalies are a set of samples that do not follow the normal behavior of the majority of data. In an industrial dataset, anomalies appear in a very small number of samples. Currently, deep learning-based models have achieved important advances in image anomaly detection. However, with general models, real-world application data consisting of non-ideal images, also known as poison images, become a challenge. When the work environment is not conducive to consistently acquiring a good or ideal sample, an additional adaptive learning model is needed. In this work, we design a potential methodology to tackle poison or non-ideal images that commonly appear in industrial production lines by enhancing the existing training data. We propose Hierarchical Image Transformation and Multi-level Features (HIT-MiLF) modules for an anomaly detection network to adapt to perturbances from novelties in testing images. This approach provides a hierarchical process for image transformation during pre-processing and explores the most efficient layer of extracted features from a CNN backbone. The model generates new transformations of training samples that simulate the non-ideal condition and learn the normality in high-dimensional features before applying a Gaussian mixture model to detect the anomalies from new data that it has never seen before. Our experimental results show that hierarchical transformation and multi-level feature exploration improve the baseline performance on industrial metal datasets.

Ivermectin's Role in the Prevention of COVID-19: A Systematic Review and Meta-Analysis.

This systematic review was performed to determine the population that benefited from prophylactic ivermectin. Seven databases of health-related studies were searched for eligible trials without language restrictions. Randomized controlled trials (RCTs) and cohort studies investigating ivermectin for coronavirus disease 2019 (COVID-19) prevention were included. Data were pooled using a random-effects model, and subgroups were analyzed by study type and the pre- or postexposure population. The certainty of the evidence was determined by the Grading of Recommendations Assessment, Development, and Evaluation approach. Furthermore, 4 RCTs and 4 cohort studies with a moderate to high risk of bias were included in the analysis. The prophylactic use of ivermectin significantly decreased the overall incidence of COVID-19 (odds ratio [OR], 0.26; 95% confidence interval [CI], 0.16-0.44). Nevertheless, the positive result was not supported by the RCT. Ivermectin was associated with a lower risk of COVID-19 (OR, 0.22; 95% CI, 0.12-0.40) in the preexposure population, whereas no protective effect was observed in the postexposure population (OR, 0.39; 95% CI, 0.09-1.67). In summary, prophylactic ivermectin did not prevent COVID-19 in the postexposure population. Although the protective effect of ivermectin was shown in the overall and preexposure populations, the results were unreliable owing to poor-quality evidence.

Specific Expert Learning: Enriching Ensemble Diversity via Knowledge Distillation.

In recent years, ensemble methods have shown sterling performance and gained popularity in visual tasks. However, the performance of an ensemble is limited by the paucity of diversity among the models. Thus, to enrich the diversity of the ensemble, we present the distillation approach-learning from experts (LFEs). Such method involves a novel knowledge distillation (KD) method that we present, specific expert learning (SEL), which can reduce class selectivity and improve the performance on specific weaker classes and overall accuracy. Through SEL, models can acquire different knowledge from distinct networks with various areas of expertise, and a highly diverse ensemble can be obtained afterward. Our experimental results demonstrate that, on CIFAR-10, the accuracy of the ResNet-32 increases 0.91% with SEL, and that the ensemble trained by SEL increases accuracy by 1.13%. Compared to state-of-the-art approaches, for example, DML only improves accuracy by 0.3% and 1.02% on single ResNet-32 and the ensemble, respectively. Furthermore, our proposed architecture also can be applied to ensemble distillation (ED), which applies KD on the ensemble model. In conclusion, our experimental results show that our proposed SEL not only improves the accuracy of a single classifier but also boosts the diversity of the ensemble model.

Visfatin-Induced Inhibition of miR-1264 Facilitates PDGF-C Synthesis in Chondrosarcoma Cells and Enhances Endothelial Progenitor Cell Angiogenesis.

New treatments for chondrosarcoma are extremely important. Chondrosarcoma is a primary malignant bone tumor with a very unfavorable prognosis. High-grade chondrosarcoma has a high potential to metastasize to any organ in the body. Platelet-derived growth factor (PDGF) is a potent angiogenic factor that promotes tumor angiogenesis and metastasis. The adipocytokine visfatin promotes metastatic potential of chondrosarcoma; however, the role of visfatin in angiogenesis in human chondrosarcoma is unclear. We report that the levels of PDGF-C expression were positively correlated with tumor stages, significantly higher than the levels of expression in normal cartilage. Visfatin increased PDGF-C expression and endothelial progenitor cell (EPC) angiogenesis through the PI3K/Akt/mTOR signaling pathway, and dose-dependently down-regulated the synthesis of miR-1264, which targets the 3'-UTR of PDGF-C. Additionally, we discovered inhibition of visfatin or PDGF-C in chondrosarcoma tumors significantly reduced tumor angiogenesis and size. Our results indicate that visfatin inhibits miR-1264 production through the PI3K/Akt/mTOR signaling cascade, and thereby promotes PDGF-C expression and chondrosarcoma angiogenesis. Visfatin may be worth targeting in the treatment of chondrosarcoma angiogenesis.

Mini Review: Molecular Interpretation of the IGF/IGF-1R Axis in Cancer Treatment and Stem Cells-Based Therapy in Regenerative Medicine.

In addition to the fundamental role of insulin-like growth factor (IGF)/IGF-1 receptor (IGF-1R) signaling dysregulation in cancer initiation and proliferation, the IGF/IGF-1R signaling also plays an important role in the maintenance of stem cell characteristics and enhancement of stem cell-based therapeutic efficacy. This review focused on the role of IGF/IGF-1R signaling in preclinical IGF-targeted therapies, including IGF-1R monoclonal antibodies, IGF-1R tyrosine kinase inhibitors, and neutralizing antibodies of IGFs in multiple tumors and endocrine disorders. On the other hand, the function of IGF/IGF-1R signaling in stem cell self-renewal, pluripotency and therapeutic efficacy in regenerative medicine was outlined. Finally, the review summarized ongoing studies on IGF/IGF-1R signaling blockade in multiple cancers and highlighted the IGF-1R signaling modifications in stem cells as a potential strategy to improve stem cell-based therapeutics in regenerative medicine.

Novel Spatio-Temporal Continuous Sign Language Recognition Using an Attentive Multi-Feature Network.

Given video streams, we aim to correctly detect unsegmented signs related to continuous sign language recognition (CSLR). Despite the increase in proposed deep learning methods in this area, most of them mainly focus on using only an RGB feature, either the full-frame image or details of hands and face. The scarcity of information for the CSLR training process heavily constrains the capability to learn multiple features using the video input frames. Moreover, exploiting all frames in a video for the CSLR task could lead to suboptimal performance since each frame contains a different level of information, including main features in the inferencing of noise. Therefore, we propose novel spatio-temporal continuous sign language recognition using the attentive multi-feature network to enhance CSLR by providing extra keypoint features. In addition, we exploit the attention layer in the spatial and temporal modules to simultaneously emphasize multiple important features. Experimental results from both CSLR datasets demonstrate that the proposed method achieves superior performance in comparison with current state-of-the-art methods by 0.76 and 20.56 for the WER score on CSL and PHOENIX datasets, respectively.

A deep feature-level fusion model for masked face identity recommendation system.

The widespread occurrences of airborne outbreaks (e.g., COVID-19) and pollution (e.g., PM2.5) have urged people in the affected regions to protect themselves by wearing face masks. In certain areas, wearing masks amidst such health-endangering times is even enforced by law. While most people wear masks to guard themselves against airborne substances, some exploit such excuses and use face masks to conceal their identity for criminal purposes such as shoplifting, robbery, drug transport, and assault. While automatic face recognition models have been proposed, most of these models aim to identify clear, unobstructed faces for authentication purposes and cannot effectively handle cases where masks cover most facial areas. To mitigate such a problem, this paper proposes a deep-learning-based feature-fusion framework, FIREC, that combines additional demographic-estimated features such as age, gender, and race into the underlying facial representation to compensate for the information lost due to mask obstruction. Given an image of a masked face, our system recommends a ranked list of potential identities of the person behind the mask. Empirical results show that the best configuration of our proposed framework can recognize bare faces and masked faces with the accuracy of 99.34% and 97.65% in terms of Hit@10, respectively. The proposed framework could greatly benefit high-recall facial identity recognition applications such as identifying potential suspects from CCTV or passers-by's cameras, especially during crisis times when people commonly cover their faces with protective masks.

IL-17 Facilitates VCAM-1 Production and Monocyte Adhesion in Osteoarthritis Synovial Fibroblasts by Suppressing miR-5701 Synthesis.

Osteoarthritis (OA) is characterized by the infiltration and adhesion of monocytes into the inflamed joint synovium. Interleukin (IL)-17 is a critical inflammatory mediator that participates in the progression of OA, although the mechanisms linking IL-17 and monocyte infiltration are not well understood. Our analysis of synovial tissue samples retrieved from the Gene Expression Omnibus (GEO) dataset exhibited higher monocyte marker (CD11b) and vascular cell adhesion molecule 1 (VCAM-1) levels in OA samples than in normal, healthy samples. The stimulation of human OA synovial fibroblasts (OASFs) with IL-17 increased VCAM-1 production and subsequently enhanced monocyte adhesion. IL-17 affected VCAM-1-dependent monocyte adhesion by reducing miR-5701 expression through the protein kinase C (PKC)-α and c-Jun N-terminal kinase (JNK) signaling cascades. Our findings improve our understanding about the effect of IL-17 on OA progression and, in particular, VCAM-1 production and monocyte adhesion, which may help with the design of more effective OA treatments.

Enhancing Precision with an Ensemble Generative Adversarial Network for Steel Surface Defect Detectors (EnsGAN-SDD).

Defects are the primary problem affecting steel product quality in the steel industry. The specific challenges in developing detect defectors involve the vagueness and tiny size of defects. To solve these problems, we propose incorporating super-resolution technique, sequential feature pyramid network, and boundary localization. Initially, the ensemble of enhanced super-resolution generative adversarial networks (ESRGAN) was proposed for the preprocessing stage to generate a more detailed contour of the original steel image. Next, in the detector section, the latest state-of-the-art feature pyramid network, known as De-tectoRS, utilized the recursive feature pyramid network technique to extract deeper multi-scale steel features by learning the feedback from the sequential feature pyramid network. Finally, Side-Aware Boundary Localization was used to precisely generate the output prediction of the defect detectors. We named our approach EnsGAN-SDD. Extensive experimental studies showed that the proposed methods improved the defect detector's performance, which also surpassed the accuracy of state-of-the-art methods. Moreover, the proposed EnsGAN achieved better performance and effectiveness in processing time compared with the original ESRGAN. We believe our innovation could significantly contribute to improved production quality in the steel industry.